To follow up on yesterday’s anticipated financial earnings reports from Meta, Microsoft, Alphabet and Amazon, we see that spending as a percentage of revenue was significantly higher than industry averages. For example, Meta’s capex intensity was shown to be 2.4x higher than the software industry average. As mentioned yesterday, the four companies have committed more than $600 billion in spend for data centers, power, chips, and other AI infrastructure. So how did they fare?

Each showed a general trend of beating revenue and profit expectations, but a continued focus on elevated Capex guidance:

Alphabet (GOOGL):

Revenue: $109.9 billion (+22% YoY).
CapEx: $35.7 billion in Q1.

Shares spiked roughly 6%, with earnings of $2.81 per share outpacing the $2.64 projected on +22% revenue growth to $109.9 billion (minus Traffic Acquisition Costs, as Zacks reported it, with a figure reaching $96.2 billion, ahead of the $92.2 billion expected). Analysts praised the company’s Cloud backlog ($460 billion+), and its stance as an “AI landlord,” delivering TPU chips to customers – a potential $900 billion business. Citi and other firms noted that Alphabet’s cloud margins nearly doubled, providing a clearer path to profitability, despite its elevated spending.

Meta:

Revenue: $56.3 billion (+33% YoY).
CapEx: $19.84 billion in Q1.

Meta beat the $6.71 estimate and +62% YoY on $56.31 billion in revenues, surpassing the $55.49 billion expected, +33% YoY. Despite that, Meta’s stock fell about 9% after the company raised its 2026 CapEx guide to $125–$145 billion. This spooked the market a bit, because its Capex is growing faster than its revenue, with some viewing its Superintelligence Labs as another metaverse-like gamble in the AI arms race.

Microsoft:

Revenue: $82.89 billion (+18% YoY).
CapEx: $34.9 billion in Q1.

Microsoft revenues of $82.89 billion easily surpassed the $81.40 billion anticipated, but the company experienced a slight stock dip due to what may have been seen as mixed execution, with Microsoft projecting $190 billion in spend, but admitting it will be “capacity constrained” through 2026. Overall, high price targets remained, showing confidence the backlog will eventually convert to high-margin profit.

Amazon:

Revenue: $181.5 billion (+17% YoY).
CapEx: Not fully broken out for Q1, but hinted at very high spending, comparable to Alphabet.

Some say Amazon “blew the doors off” profit expectations, particularly in AWS which saw its fastest growth in 15 quarters (28%). Despite the positive results, shares initially slipped slightly in after-hours trading due to high Capex to support its 2.1 million+ AI chip deployment.

Susana Schwartz
Technology Editor
RCRTech

AI Infrastructure Top Stories

Growing interest in ‘microgrids’: ABB’s Sébastien Surply says “With AI data centers now requiring 100+ MW of continuous power,” Microgrids are being considered as a resilient, low-carbon on-site energy alternatives in data centers.

UK wants skin in the game: Though the U.S. and Taiwan lead, the U.K. government’s new national AI Hardware Plan aims to capture “at least 5%” of what Tech Secretary Liz Kendall estimates will be a $1T semiconductor market by 2030s.

Telcos pursuing AI infra: Like China Mobile and other Chinese telcos, ZTE is expanding into AI infrastructure to bolster relationships and growth with companies like Nvidia, Intel, Alibaba, ByteDance and others, according to Omdia’s Guang Yang.

View More News

AI Today: What You Need to Know

Oklahoma data center proposed: City leaders in Broken Arrow have been approached about a predevelopment meeting for a potential data center on the city’s east side. This comes as OK lawmakers consider rising community opposition.

TSMC’s explosive growth: Taiwan Semiconductor Manufacturing Company (TSMC) reported strong Q1 2026 earnings, with 35.1% YoY revenue growth, and a 58% increase in Q1 profit, driven by 3nm and advanced packaging for AI chips.

NVIDIA new models: Nvidia’s newly launched Nemotron 3 Nano Omni Model offers 9x improvement in reasoning for high efficiency and accuracy in agentic workflows, such as computer use, document intelligence and audio-video reasoning.

Vertiv acquires STL & others: In acquiring Strategic Thermal Labs, a specialist in direct-to-chip and cold-plate designso, Vertiv follows up on its March acquisition of ThermoKey and April deal for BMarko Structures – all for thermal management.

1-Bit LLM breakthrough: 1-Bit LLMs, pioneered by startups like PrismML’s Bonsai 8B LM, are about 14x smaller than a standard 16-bit model and drastically reduce energy consumption (up to, 100x), enabling high-performance AI on edge devices.

OpenAI-xAI trial continues: Elon Musk will take the stand again today, following yesterday’s heated, multi-hour exchange with OpenAI lawyer William Savitt. Once Musk is done, his attorneys will call Jared Birchall and others to the testify.

RCR Events

Defense Communications Forum, April 28th

Join defense leaders, technology innovators and system integrators and explore how commercial network solutions can be adapted to meet defense-grade requirements, enabling secure, resilient, AI-driven, multi-bearer networks.

Test & Measurement Forum, May 19

Join the annual Test & Measurement Forum where industry leaders, innovators, and engineers come together to tackle the most pressing test and measurement challenges in telco today. Register now

Telco AI Forum, June 16th

Telco AI Forum brings together operators, vendors, hyperscalers, and academia to explore how the industry and partnership ecosystem evolution are laying the foundations for AI-native 6G networks and unlocking ROI. Register now